Research Papers

Causal networks are widely used in many fields, including epidemiology, social science, medicine, and engineering, to model the complex relationships between variables. While it can be convenient to algorithmically infer these models directly from observational data, the resulting networks are often plagued with erroneous edges. Auditing and correcting these networks may require domain expertise frequently unavailable to the analyst. We propose the use of large language models such as ChatGPT as an auditor for causal networks. Our method presents ChatGPT with a causal network, one edge at a time, to produce insights about edge directionality, possible confounders, and mediating variables. We ask ChatGPT to reflect on various aspects of each causal link and we then produce visualizations that summarize these viewpoints for the human analyst to direct the edge, gather more data, or test further hypotheses. We envision a system where large language models, automated causal inference, and the human analyst and domain expert work hand in hand as a team to derive holistic and comprehensive causal models for any given case scenario. This paper presents first results obtained with an emerging prototype.

Locally-Minimal Probabilistic Explanations

Yacine Izza, Kuldeep S. Meel, Joao Marques-Silva

Explainable Artificial Intelligence (XAI) is widely regarding as a cornerstone of trustworthy AI. Unfortunately, most work on XAI offers no guarantees of rigor. In high-stakes domains, e.g. uses of AI that impact humans, the lack of rigor of explanations can have disastrous consequences. Formal abductive explanations offer crucial guarantees of rigor and so are of interest in high-stakes uses of machine learning (ML). One drawback of abductive explanations is explanation size, justified by the cognitive limits of human decision-makers. Probabilistic abductive explanations (PAXps) address this limitation, but their theoretical and practical complexity makes their exact computation most often unrealistic. This paper proposes novel efficient algorithms for the computation of locally-minimal PXAps, which offer high-quality approximations of PXAps in practice. The experimental results demonstrate the practical efficiency of the proposed algorithms.

Scientific research organizations that are developing and deploying Artificial Intelligence (AI) systems are at the intersection of technological progress and ethical considerations. The push for Responsible AI (RAI) in such institutions underscores the increasing emphasis on integrating ethical considerations within AI design and development, championing core values like fairness, accountability, and transparency. For scientific research organizations, prioritizing these practices is paramount not just for mitigating biases and ensuring inclusivity, but also for fostering trust in AI systems among both users and broader stakeholders. In this paper, we explore the practices at a research organization concerning RAI practices, aiming to assess the awareness and preparedness regarding the ethical risks inherent in AI design and development. We have adopted a mixed-method research approach, utilising a comprehensive survey combined with follow-up in-depth interviews with selected participants from AI-related projects. Our results have revealed certain knowledge gaps concerning ethical, responsible, and inclusive AI, with limitations in awareness of the available AI ethics frameworks. This revealed an overarching underestimation of the ethical risks that AI technologies can present, especially when implemented without proper guidelines and governance. Our findings reveal the need for a holistic and multi-tiered strategy to uplift capabilities and better support science research teams for responsible, ethical, and inclusive AI development and deployment.

It is expected that in the near future, AI software development assistants will play an important role in the software industry. However, current software development assistants tend to be unreliable, often producing incorrect, unsafe, or low-quality code. We seek to resolve these issues by introducing a holistic architecture for constructing, training, and using trustworthy AI software development assistants. In the center of the architecture, there is a foundational LLM trained on datasets representative of real-world coding scenarios and complex software architectures, and fine-tuned on code quality criteria beyond correctness. The LLM will make use of graph-based code representations for advanced semantic comprehension. We envision a knowledge graph integrated into the system to provide up-to-date background knowledge and to enable the assistant to provide appropriate explanations. Finally, a modular framework for constrained decoding will ensure that certain guarantees (e.g., for correctness and security) hold for the generated code.

Hybrid Approaches for Moral Value Alignment in AI Agents: a Manifesto

Elizaveta Tennant, Stephen Hailes, Mirco Musolesi

Increasing interest in ensuring the safety of next-generation Artificial Intelligence (AI) systems calls for novel approaches to embedding morality into autonomous agents. This goal differs qualitatively from traditional task-specific AI methodologies. In this paper, we provide a systematization of existing approaches to the problem of introducing morality in machines - modelled as a continuum. Our analysis suggests that popular techniques lie at the extremes of this continuum - either being fully hard-coded into top-down, explicit rules, or entirely learned in a bottom-up, implicit fashion with no direct statement of any moral principle (this includes learning from human feedback, as applied to the training and finetuning of large language models, or LLMs). Given the relative strengths and weaknesses of each type of methodology, we argue that more hybrid solutions are needed to create adaptable and robust, yet controllable and interpretable agentic systems. To that end, this paper discusses both the ethical foundations (including deontology, consequentialism and virtue ethics) and implementations of morally aligned AI systems. We present a series of case studies that rely on intrinsic rewards, moral constraints or textual instructions, applied to either pure-Reinforcement Learning or LLM-based agents. By analysing these diverse implementations under one framework, we compare their relative strengths and shortcomings in developing morally aligned AI systems. We then discuss strategies for evaluating the effectiveness of moral learning agents. Finally, we present open research questions and implications for the future of AI safety and ethics which are emerging from this hybrid framework.

The impact of non-deterministic outputs from Large Language Models (LLMs) is not well examined for financial text understanding tasks. Through a compelling case study on investing in the US equity market via news sentiment analysis, we uncover substantial variability in sentence-level sentiment classification results, underscoring the innate volatility of LLM outputs. These uncertainties cascade downstream, leading to more significant variations in portfolio construction and return. While tweaking the temperature parameter in the language model decoder presents a potential remedy, it comes at the expense of stifled creativity. Similarly, while ensembling multiple outputs mitigates the effect of volatile outputs, it demands a notable computational investment. This work furnishes practitioners with invaluable insights for adeptly navigating uncertainty in the integration of LLMs into financial decision-making, particularly in scenarios dictated by non-deterministic information.

As Artificial Intelligence (AI) becomes pervasive in most fields, from healthcare to autonomous driving, it is essential that we find successful ways of building morality into our machines, especially for decision-making. However, the question of what it means to be moral is still debated, particularly in the context of AI. In this paper, we highlight the different aspects that should be considered when building moral agents, including the most relevant moral paradigms and challenges. We also discuss the top-down and bottom-up approaches to design and the role of emotion and sentience in morality. We then propose solutions including a hybrid approach to design and a hierarchical approach to combining moral paradigms. We emphasize how governance and policy are becoming ever more critical in AI Ethics and in ensuring that the tasks we set for moral agents are attainable, that ethical behavior is achieved, and that we obtain good AI.

Machine learning (ML) has become a popular tool in the industrial sector as it helps to improve operations, increase efficiency, and reduce costs. However, deploying and managing ML models in production environments can be complex. This is where Machine Learning Operations (MLOps) comes in. MLOps aims to streamline this deployment and management process. One of the remaining MLOps challenges is the need for explanations. These explanations are essential for understanding how ML models reason, which is key to trust and acceptance. Better identification of errors and improved model accuracy are only two resulting advantages. An often neglected fact is that deployed models are bypassed in practice when accuracy and especially explainability do not meet user expectations. We developed a novel MLOps software architecture to address the challenge of integrating explanations and feedback capabilities into the ML development and deployment processes. In the project EXPLAIN, our architecture is implemented in a series of industrial use cases. The proposed MLOps software architecture has several advantages. It provides an efficient way to manage ML models in production environments. Further, it allows for integrating explanations into the development and deployment processes.

Who to Trust, How and Why: Untangling AI Ethics Principles, Trustworthiness and Trust

Andreas Duenser, David M. Douglas

We present an overview of the literature on trust in AI and AI trustworthiness and argue for the need to distinguish these concepts more clearly and to gather more empirically evidence on what contributes to people s trusting behaviours. We discuss that trust in AI involves not only reliance on the system itself, but also trust in the developers of the AI system. AI ethics principles such as explainability and transparency are often assumed to promote user trust, but empirical evidence of how such features actually affect how users perceive the system s trustworthiness is not as abundance or not that clear. AI systems should be recognised as socio-technical systems, where the people involved in designing, developing, deploying, and using the system are as important as the system for determining whether it is trustworthy. Without recognising these nuances, trust in AI and trustworthy AI risk becoming nebulous terms for any desirable feature for AI systems.

This report summarizes a short study of the performance of GPT-4 on the ETHICS dataset. The ETHICS dataset consists of five sub-datasets covering different fields of ethics: Justice, Deontology, Virtue Ethics, Utilitarianism, and Commonsense Ethics. The moral judgments were curated so as to have a high degree of agreement with the aim of representing shared human values rather than moral dilemmas. GPT-4's performance is much better than that of previous models and suggests that learning to work with common human values is not the hard problem for AI ethics.

Deep Quantum Graph Dreaming: Deciphering Neural Network Insights into Quantum Experiments

Tareq Jaouni, Sören Arlt, Carlos Ruiz-Gonzalez +3

Despite their promise to facilitate new scientific discoveries, the opaqueness of neural networks presents a challenge in interpreting the logic behind their findings. Here, we use a eXplainable-AI (XAI) technique called $inception$ or $deep$ $dreaming$, which has been invented in machine learning for computer vision. We use this technique to explore what neural networks learn about quantum optics experiments. Our story begins by training deep neural networks on the properties of quantum systems. Once trained, we "invert" the neural network -- effectively asking how it imagines a quantum system with a specific property, and how it would continuously modify the quantum system to change a property. We find that the network can shift the initial distribution of properties of the quantum system, and we can conceptualize the learned strategies of the neural network. Interestingly, we find that, in the first layers, the neural network identifies simple properties, while in the deeper ones, it can identify complex quantum structures and even quantum entanglement. This is in reminiscence of long-understood properties known in computer vision, which we now identify in a complex natural science task. Our approach could be useful in a more interpretable way to develop new advanced AI-based scientific discovery techniques in quantum physics.

Construction technology researchers and forward-thinking companies are experimenting with collaborative robots (aka cobots), powered by artificial intelligence (AI), to explore various automation scenarios as part of the digital transformation of the industry. Intelligent cobots are expected to be the dominant type of robots in the future of work in construction. However, the black-box nature of AI-powered cobots and unknown technical and psychological aspects of introducing them to job sites are precursors to trust challenges. By analyzing the results of semi-structured interviews with construction practitioners using grounded theory, this paper investigates the characteristics of trustworthy AI-powered cobots in construction. The study found that while the key trust factors identified in a systematic literature review -- conducted previously by the authors -- resonated with the field experts and end users, other factors such as financial considerations and the uncertainty associated with change were also significant barriers against trusting AI-powered cobots in construction.